{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"pygments_lexer":"ipython3","nbconvert_exporter":"python","version":"3.6.4","file_extension":".py","codemirror_mode":{"name":"ipython","version":3},"name":"python","mimetype":"text/x-python"},"kaggle":{"accelerator":"none","dataSources":[{"sourceId":3960,"databundleVersionId":868348,"sourceType":"competition"}],"isInternetEnabled":true,"language":"python","sourceType":"notebook","isGpuEnabled":false}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"markdown","source":"# AES - AdaCore Logic Based\n\n## Overview\n\nThis notebook explores a deterministic, logic-based approach to seizure prediction\nusing intracranial EEG (iEEG) time-series data.\n\nThe method does **not** rely on machine learning, templates, or learned models.\nInstead, it focuses on detecting *structural irregularities* in noisy signals\nthrough a set of explicit, explainable rules.\n\nThe primary goal is not clinical deployment,\nbut to evaluate whether a rule-based signal detection framework\ncan generalize beyond its original domain\nand remain stable under unseen conditions.\n\n---\n\n## Motivation\n\nIn many scientific time-series problems,\nthe exact shape of a meaningful signal is unknown.\n\nThis is particularly true in medical and physiological data,\nwhere variability between subjects is large\nand noise dominates most observations.\n\nRather than attempting to model *what a seizure looks like*,\nthis approach focuses on a different question:\n\n> **Where does the signal behave differently from its local background?**\n\nBy emphasizing local change and consistency,\nthe method aims to remain robust\neven when signal morphology varies significantly.\n\n---\n\n## High-Level Approach\n\nThe detection pipeline is built on three principles:\n\n1. **Local change detection**  \n   Sudden structural changes are more informative than absolute amplitude.\n\n2. **Robust normalization**  \n   All scores are normalized against recent local statistics\n   to suppress outlier explosions.\n\n3. **Multi-view stabilization**  \n   The same logic is evaluated under multiple slightly different configurations,\n   and combined conservatively.\n\nThis results in a fully deterministic pipeline\nwhose behavior can be inspected and audited at every stage.\n\n---\n\n## Feature Construction\n\nRaw EEG signals are first divided into multiple frequency bands.\n\nWithin each band, simple windowed statistics\n(such as RMS-like measures)\nare computed over sliding time windows.\n\nThe result is a compact time-series representation\nthat preserves temporal structure\nwhile reducing sensitivity to raw waveform noise.\n\nNo assumptions are made about waveform shape,\nphase alignment, or specific frequency signatures.\n\n---\n\n## Anchor-Based Scanning\n\nThe resulting time-series is scanned at **every frame**.\n\nThis avoids parity-related blind spots\nthat can occur when scanning with larger strides,\nwhere true peaks may be skipped entirely.\n\nAt each frame, the local change is computed\nas the absolute difference from the previous frame.\n\nThese local differences form a new sequence\nthat highlights potential “anchors”\nwhere the signal behaves unusually.\n\n---\n\n## Robust Scoring\n\nRaw local changes can vary by several orders of magnitude,\nespecially in unseen test data.\n\nTo stabilize the score distribution:\n\n- Each local change is normalized\n  by the median of recent changes\n  in a trailing window.\n- A logarithmic compression (`log1p`) is applied\n  to suppress extreme outliers\n  while preserving relative ordering.\n\nThis produces a score that behaves\nlike a local signal-to-noise indicator,\nbut remains numerically stable.\n\n---\n\n## From Anchors to a Sample Score\n\nFor each sample, the strongest anchor candidates are selected.\n\nRather than relying on a single maximum,\nthe final score is computed\nfrom a small set of top anchors.\n\nThis reduces sensitivity to accidental spikes\nand emphasizes consistent local structure.\n\n---\n\n## Multi-View Evaluation\n\nTo further improve robustness,\nthe entire process is repeated\nunder multiple slightly different configurations (“views”).\n\nThese views differ in parameters such as\nfrequency band composition or windowing details.\n\nThe final prediction is obtained\nby taking the **median** score across all views.\n\nThis conservative aggregation strategy\ntrades a small amount of peak performance\nfor significantly improved stability.\n\n---\n\n## Why This Matters\n\nThis notebook demonstrates that:\n\n- Deterministic, rule-based methods\n  can remain competitive in noisy, real-world data.\n- Explicit logic allows detailed inspection and auditing,\n  which is especially important in medical-related domains.\n- Signal detection can be framed around *change*\n  rather than *shape*.\n\nThe same framework can be applied\nto other time-series problems\nwhere signals are rare, ambiguous, or poorly defined.\n\n---\n\n## Notes on Safety and Interpretation\n\nAlthough this work uses medical data,\nit is strictly a technical exploration.\n\nAll outputs should be interpreted\nas experimental signal scores,\nnot medical diagnoses or clinical recommendations.\n\nThe emphasis on explicit logic and logging\nis intentional,\nreflecting the need for traceability and accountability\nwhen working with medically adjacent data.\n\n---\n\n## Result\n\n**Submission Status:** *After deadline (evaluation unavailable)*\n\nThe submission was completed and uploaded successfully.  \nBecause the competition deadline had already passed,  \nno leaderboard score was produced.\n\nThis notebook therefore focuses on demonstrating  \nthe generality and robustness of the deterministic, template-free pipeline,  \nrather than optimizing for a specific leaderboard score.\n\nA screenshot of the submission status is shown below.\n\n![ss.png](attachment:55afa20b-19f6-409e-86f5-64fa45e9d5fe.png)\n\n### Next Steps\n\nThis work serves as a proof of generality for a deterministic,  \ntemplate-free signal detection pipeline.\n\nPossible next steps include:\n\n- Applying the same framework to other active time-series competitions  \n  to obtain comparable leaderboard scores.\n- Exploring minor refinements in view fusion or normalization,  \n  while keeping the core logic unchanged.\n- Using this pipeline as a reference implementation  \n  for explainable, non-ML signal detection tasks.\n\nFurther exploration will depend on the choice of the next target dataset.\n\n---\n\n## Closing Remarks\n\nWhen the signal is unknown,\nit can be more effective to focus on *how the signal changes*\nrather than *what the signal looks like*.\n\nThis notebook presents one such approach,\ngrounded in explicit logic,\nrobust normalization,\nand conservative aggregation.","metadata":{},"attachments":{"55afa20b-19f6-409e-86f5-64fa45e9d5fe.png":{"image/png":"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"}}}]}